Predicting Enrollment: 6 Tips for Sites to Estimate the Right Number

Kristina Lopienski
January 12th, 2017

Estimating enrollment is a typical part of the feasibility process for any clinical trial, but defining targets and accurately predicting numbers is a bit of a guessing game. In choosing the right number, sites face the fear of under-estimating and risks of over-estimating. The reality is, predicting enrollment is always just that–an estimate. Sponsors and CROs make decisions regarding enrollment potential and the number of sites based on preliminary site estimates, which rarely come to fruition. Most trials end up doubling their original timelines to meet enrollment goals, and 48% of sites under-enroll study volunteers (Tufts, 2013).

To improve these percentages, we share an excerpt from the free eBook, “Improving Site-Sponsor Relationships: Proactive Strategies for Transparent Clinical Trials.” Here are a few ways sites can improve their enrollment estimates.

1. Give a thoughtful answer.

Predicting enrollment can be a guessing game–more of an art than a science–but sites must do as much as possible to be accurate. Consider questions such as, is the indication feasible for your site? Do you have the population to draw from? Look at a variety of data sources to learn precedence of the disease. For some indications, it is important to know whether the PI has a relationship with the patients. The type of indication will have significant impact on recruitment methods. Look at your database and the metrics that can be extracted. Of the full patient population, how many would actually want to be involved in this type of trial? If you once had a huge patient pool, have those patients since gone on treatments? Communicate with patients and staff to get a better sense of what to estimate.

2. Be upfront with the sponsor about competing studies.

It would be problematic to take on a competing study, as it can interfere with enrollment. If you decline a study for this reason and it’s an indication that you are typically very good at, tell the sponsor why you are declining. This will help build a good relationship and let the sponsors know that future studies in the same indication would be something you are interested in conducting.

3. Don’t overestimate your enrollment forecast.

Be realistic and conservative in your estimate–sponsors and CROs would rather have sites underestimate than overestimate. Inflated enrollment estimates from sites are a huge problem for sponsors/CROs. If sites only enroll half the patients they predicted, the sponsor/CRO has to quickly add sites to increase enrollment. Also, if a site’s estimate is much higher than the typical site response, there is often doubt or need for further clarification. Honesty will get sites a long way with the sponsor. If a site is honest with the sponsor when they cannot do the study or when they are not set up for it at the moment, the sponsor may still consider the site for future studies.

4. Showcase your past enrollment performance.

As recruitment potential is often a key factor driving the site selection decision, provide an enrollment analysis of prior performance on similar studies. This “report card” supplements the information provided in the feasibility questionnaire and demonstrates how consistently a site has performed in terms of their ability to achieve timely First Patient In (FPI), as well as overall enrollment targets. Many sponsors have indicated that, if a site can provide objective evidence of performance, they trust the reliability of numbers the site provides is much greater than if the site had submitted an estimate alone.

5. If you aren’t selected for a study, ask the sponsor why

Reach out to the sponsor and ask them why you didn’t get selected for a certain study. The sponsor may hesitate to proactively give you reasons, but should respond, especially if you ask over the phone. With the feedback you get, be sure to use it for awareness and improvement. Don’t just increase your enrollment numbers next time so you will be selected. Being honest will always help your chance of future studies.

6. Update enrollment estimates once you have more information.

Many times the initial enrollment estimate is based off of preliminary abstracts and not a final protocol with full inclusion/exclusion criteria. It is important to update your estimates upon receiving the full set of information, so the sponsor can plan accordingly. To do this,

conduct a more detailed review of the eligibility criteria and conduct further enrollment validation activities (e.g., EMR search, sample chart reviews, etc.) to update enrollment estimates. If your numbers are reduced due to this new information, update the sponsor. You can even supplement your new information by preparing a preliminary recruitment plan to illustrate that you are thinking about supplementing your own pool of patients with recruitment from external sources.